codex

Codex Skill Guide

Running a Task

1. Ask the user (via AskUserQuestion) which model to run: gpt-5 or gpt-5-codex.
2. Ask the user (via AskUserQuestion) which reasoning effort to use: low, medium, or high.
3. Select the sandbox mode required for the task; default to --sandbox read-only unless edits or network access are necessary.
4. Assemble the command with the appropriate options:
   • -m, --model <MODEL>
   • --config model_reasoning_effort="<low|medium|high>"
   • --sandbox <read-only|workspace-write|danger-full-access>
   • --full-auto
   • -C, --cd <DIR>
   • --skip-git-repo-check
5. When continuing a previous session, use codex exec resume --last via stdin. IMPORTANT: When resuming, you CANNOT specify model, reasoning effort, or other flags—the session retains all settings from the original run. Resume syntax: echo "your prompt here" | codex exec resume --last
6. Run the command, capture stdout/stderr, and summarize the outcome for the user.

Quick Reference

Use case	Sandbox mode	Key flags
Read-only review or analysis	read-only	--sandbox read-only
Apply local edits	workspace-write	--sandbox workspace-write --full-auto
Permit network or broad access	danger-full-access	--sandbox danger-full-access --full-auto
Resume recent session	Inherited from original	echo "prompt" | codex exec resume --last (no flags allowed)
Run from another directory	Match task needs	-C <DIR> plus other flags

Following Up

• After every codex command, immediately use AskUserQuestion to confirm next steps, collect clarifications, or decide whether to resume with codex exec resume --last.
• When resuming, pipe the new prompt via stdin: echo "new prompt" | codex exec resume --last. The resumed session automatically uses the same model, reasoning effort, and sandbox mode from the original session.
• Restate the chosen model, reasoning effort, and sandbox mode when proposing follow-up actions.

Error Handling

• Stop and report failures whenever codex --version or a codex exec command exits non-zero; request direction before retrying.
• Before you use high-impact flags (--full-auto, --sandbox danger-full-access, --skip-git-repo-check) ask the user for permission using AskUserQuestion unless it was already given.
• When output includes warnings or partial results, summarize them and ask how to adjust using AskUserQuestion.